Spectroscopy and Spectral Analysis, Volume. 42, Issue 7, 2135(2022)

Experimental Research on Coal-Rock Identification Method Based on

Liang-ji XU1,*... Xue-ying MENG2,2;, Ren WEI2,2; and Kun ZHANG2,2; |Show fewer author(s)
Author Affiliations
  • 11. National Key Experiment of Mining Response and Disaster Prevention and Control in Deep Coal Mine, Huainan 232001, China
  • 22. School of Spatial Information and Geomatics Engineering, Anhui University of Science and Technology, Huainan 232001, China
  • show less
    Figures & Tables(15)
    Some coal and rock samples
    Original reflectance spectra
    Experimental chart of ash content of coal and rock samples
    Set up for determining the volatile matter in coal and gangue
    Model structure of PCA-BP neural network based on visible-near infrared spectra
    PCA-BP neural network model recognition rate
    Structure of PCA-BP neural network based on sample components
    Recognition rate of PCA-BP neural network model based on sample components
    • Table 1. Test results of industrial indicators of coal samples from two mines

      View table
      View in Article

      Table 1. Test results of industrial indicators of coal samples from two mines

      样本编号Mad/%Vad/%Aad/%
      XQM11.6028.1633.74
      XQM21.7127.1629.60
      XQM31.0028.3733.66
      XQM41.5028.0727.08
      XQM51.1027.6331.38
      XQM61.2028.5731.96
      XQM71.2027.5031.51
      XQG11.4010.8485.05
      XQG21.0010.6686.11
      XQG31.4012.3986.31
      XQG41.2010.8386.44
      P2M10.9923.6740.80
      P2M20.8533.155.86
      P2M30.7932.096.61
      P2M41.0633.747.05
      P2M50.9234.2511.24
      P2M61.3226.8522.71
      P2M70.9029.1117.00
      P2M81.3825.5932.60
      P2G10.402.7197.00
      P2G20.809.2687.10
      P2G31.809.6488.46
      P2G40.707.2785.43
      P2G51.708.9689.78
      P2G62.1012.3182.57
      P2G70.8010.7584.12
      P2G80.7010.9788.08
      P2G91.036.9392.19
      P2G101.119.2088.31
    • Table 2. Oxygen percentage contents of coal and rock samples from two mines

      View table
      View in Article

      Table 2. Oxygen percentage contents of coal and rock samples from two mines

      样品编号Na2OMgOAl2O3SiO2P2O5SO3K2OCaOTiO2Fe2O3
      XQM10.370.5130.0057.100.242.370.001.931.535.97
      XQM20.340.5327.9357.740.164.291.130.192.085.62
      XQM30.220.5228.7158.420.212.970.131.191.865.76
      XQM40.360.5527.1657.340.184.661.160.672.115.82
      XQM50.330.5228.6857.920.163.950.820.091.975.56
      XQM60.390.5128.6657.080.173.360.501.661.935.76
      XQM70.300.5128.3757.880.173.771.050.501.985.48
      XQG10.170.3333.1560.680.100.160.440.001.443.53
      XQG20.360.3132.3761.220.090.170.870.001.203.42
      XQG30.440.3135.6758.630.100.090.160.001.083.53
      XQG40.280.3232.5860.480.100.161.080.001.463.55
      P2M10.330.4328.0460.240.182.090.971.411.314.99
      P2M21.490.9013.6447.825.637.470.5611.731.808.97
      P2M31.530.9917.7657.141.159.360.000.662.808.60
      P2M41.240.9915.9955.321.7010.350.001.783.049.58
      P2M50.652.579.7138.280.235.710.0033.710.858.30
      P2M60.660.5623.4652.590.178.652.243.651.866.17
      P2M70.900.6224.2351.890.196.480.986.711.916.10
      P2M80.330.5227.1856.610.263.531.063.001.565.97
      P2G10.050.3219.9968.380.100.136.010.771.113.15
      P2G20.190.3129.6661.640.100.233.510.001.303.08
      P2G30.200.3130.1362.910.100.121.550.001.263.42
      P2G40.210.3130.2362.870.100.121.520.001.233.42
      P2G50.310.3028.8162.820.090.242.780.071.253.34
      P2G60.510.3134.5458.910.100.150.730.001.513.26
      P2G70.360.3029.8061.570.090.542.910.001.263.18
      P2G80.130.3027.8263.560.100.181.610.561.064.69
      P2G90.370.2929.4061.340.090.104.280.001.003.15
      P2G100.280.2928.2963.860.090.402.340.010.923.52
    • Table 3. Model accuracy of PCA-SVM model based on visible-near infrared spectra

      View table
      View in Article

      Table 3. Model accuracy of PCA-SVM model based on visible-near infrared spectra

      组别最优惩罚
      参数c
      方差参数
      g
      建模精度
      /%
      验证精度
      /%
      10.378 936.964 48544.44
      20.51.148 77575
      31.071 836.758 310066.67
      412.143 57555.56
      50.000 976 560.000 976 566522.22
      61.624 50.378 937555.56
      713.928 849033.33
      80.933 0312.125 79022.22
      91.231 127.857 69566.67
      1019.698 31.741 19566.67
      11675.588 10.435 289566.67
      120.000 976 560.000 976 565533.33
      131.148 7315.17310033.33
      1427.857 62.6399044.44
      150.000 976 560.000 976 565544.44
      160.000 976 5636.758 38544.44
      173.2490.203 066522.22
      181.319 548.502 99044.44
      191.231 184.448 59544.44
      2073.516 76.062 910033.33
    • Table 4. Accuracy of KPCA-SVM model

      View table
      View in Article

      Table 4. Accuracy of KPCA-SVM model

      组别最优惩罚
      参数c
      方差参数
      g
      建模精度
      /%
      验证精度
      /%
      10.319 52.639100100
      20.176 780.659 759088.89
      30.267 940.590100
      40.574 350.435 2810077.78
      50.435 280.353 559577.78
      60.574 350.757 869588.89
      70.203 060.659 759577.78
      834.296 84100100
      90.233 260.757 8685100
      100.000 976 562.462 3100100
      112.828 41.414 2100100
      120.267 943.2499588.89
      130.287 171.148 710066.67
      140.233 260.812 259588.89
      150.250.615 579577.78
      163.482 20.574 3590100
      170.307 793.031 49588.89
      18548.7481.866 1100100
      190.52.6399588.89
      200.203 060.574 3595100
    • Table 5. Accuracy of principal component analysis combined with support vector machine model

      View table
      View in Article

      Table 5. Accuracy of principal component analysis combined with support vector machine model

      组别最优惩罚
      参数c
      方差参数
      g
      建模精度
      /%
      验证精度
      /%
      10.574 355.2787544.44
      210.556 10.217 647066.67
      30.757 86111.430 59044.44
      46.062 96.964 48533.33
      52561.515 79522.22
      613.928 80.066 9868066.67
      70.757 86207.936 610033.33
      81.231 1630.345 910022.22
      9137.1870.094 7327066.67
      100.000 976 562.6397066.67
      116.062 90.066 9867566.67
      120.267 940.870 55100100
      130.757 8634.296 89533.33
      146.062 92.828 49066.67
      150.707 11649044.44
      160.870 5555.715 28544.44
      1725.992 11.414 28066.67
      180.615 5751.984 28044.44
      190.870 552.462 37544.44
      200.615 571.741 17033.33
    • Table 6. Model accuracy of KPCA-SVM model based on sample components

      View table
      View in Article

      Table 6. Model accuracy of KPCA-SVM model based on sample components

      组别最优惩罚
      参数c
      方差参数
      g
      建模精度
      /%
      验证精度
      /%
      10.203 060.659 7510088.89
      20.329 880.870 55100100
      30.353 550.757 86100100
      40.233 260.406 1310088.89
      50.435 280.406 1395100
      60.329 880.435 2810088.89
      70.287 170.535 8995100
      80.757 860.329 8810088.89
      924.251 50.016 746100100
      100.000 976 560.000 976 5695100
      110.000 976 560.000 976 5695100
      120.267 940.870 55100100
      130.329 880.378 9310088.89
      140.707 110.535 89100100
      1512.125 70.033 493100100
      160.000 976 560.000 976 5610088.89
      170.267 940.435 2810088.89
      180.267 940.466 5295100
      190.233 260.659 7595100
      200.267 94110077.78
    • Table 7. Algorithm model accuracy/recognition rate comparison

      View table
      View in Article

      Table 7. Algorithm model accuracy/recognition rate comparison

      特征提取算法模型建模精度
      /%
      验证精度/
      识别率/%
      可见光-
      近红外光谱
      PCA-SVM83.7566.67
      PCA-BP/46.11
      KPCA-SVM95.590.56
      样本成分含量PCA-SVM83.7550.56
      PCA-BP/46.11
      KPCA-SVM98.595
    Tools

    Get Citation

    Copy Citation Text

    Liang-ji XU, Xue-ying MENG, Ren WEI, Kun ZHANG. Experimental Research on Coal-Rock Identification Method Based on[J]. Spectroscopy and Spectral Analysis, 2022, 42(7): 2135

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Orginal Article

    Received: Jan. 11, 2021

    Accepted: --

    Published Online: Nov. 16, 2022

    The Author Email: XU Liang-ji (ljxu@aust.edu.cn)

    DOI:10.3964/j.issn.1000-0593(2022)07-2135-08

    Topics